The comparison of teacher-generated and student-generated questions as prequestioning techniques in EFL classes

ze et £ 9 S -99 0/ 'Ş % \i ^S F-^g j c Ыр р-г^ ^З ^ ıs; 3^ ii 9t ” fS 4İ? rt 'P 9'^ J *® i s '^' 1 ¡Щ ьш ш щ W ^-^Ш Ю ■ ■ -‘■ / ^^ ■ ■"5 · *^ й < ;7' ·5 ·? í " í*-^ r *C 'S' ^ ^ ν ' - · *^ ♦ » ■ w * *»"· β ' bít ζ » £ « i^ ' f''*' ' • · «/о S ' δ » Mt ΰ Ϊ . - . · <·, · '

► í9

: r,X

9 *|

j

.r:

Л

5?

'7

з^

£

TOE COMPARISON OF TEACHER-GENERATED AND STUDENT-GENERATED QUESTIONS

AS PREQUESTIONING TECHNIQUES IN EFL CLASSES

A THESIS

SUBMiriED TO THE FACULTY OF LETTER

AND THE INSTITUTE OF ECONOMICS AND SOCIAL SCIENCE OF BILKENT UNIVERSITY

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR TTIE DEGREE OF MASTER OF ARTS

IN THE TEACHING OF ENGLISH AS A FOREIGN LANGUAGE

BY

BİLGE ŞİRİN AUGUST 1992

А0(э1э .S54

W ' l

-Ь 11139

BILKENT UNIVERSIIT

INSTITUTE OF ECONOMICS AND SOCIAL SCIENCES MA THESIS E5iAMINATION RESULT FOIiM

August 31,1992

The examining committee appointed by the

In stitu te o f Economics and Social sciences fo r the th esis examination o f the MA TEFL student

. BİLGE ŞİRİN

has read the th esis o f the student. The committee has decided that the th esis

o f the student is sa tisfa c to ry .

Thesis t i t l e

Thesis Advisor

Co0imittee Members :

The conipar'ison o f teacher-generated and student-generated guestions as

prequestioning techniques in EFL

classes.

Dr. Eileen Walter

Bilkent U niversity, MA TEFL Program. Dr. Lionel Kaufman

Bilkent U niversity, MA TEFL program.

Dr. James C. Stalker

Bilkent U niversity, MA TEFL Program.

We c e r t ify that we have read th is th esis and that in our combined opinion i t i s f u ll y adequate, in scope and in

.ity, as a thesis fo r the degree o f master o f a r ts .

Eileen Walter (A dvisor) Lionel Kau^an (Committee Member, James C. Stalker Committee Member) Approved fo r the

In stitu te o f Economics and Social Sciences

ALi KARA0SMAfl06L·U Director

In stitu te of Economics and Social Science

To niy mother and fath er

TABLE OF CONTENTS

CHAPTERS PAGES

L is t oT T ables... v i i 1.0 INTRODUCTION

L. 1 Background... ...1

1.2 Goal ami Rationale o f the Study... 4

1.3 D e fin itio n s ... 6

1.4 Hypotheses...6

1.5 V a ria b le s ... 7

1.6 Overviev/ o f the Methodology... ... . . . 1

1.6.1 Subjects... .. ...8

1.6.2 Data C o lle c tio n ... 8

1.7 Data A n aly sis... 9

1.8 Expectations... 9 1.9 Limi ta tio n s... 10 1.10 Orgcuiixation o f T hesis...10 2.0 REVUÎW OF LITERATURE 2.1 Introduction... 11 2.2 P rio r Knowledge... 11 2.3 Prereading A c t iv it ie s ... 13 2.4 Prequestioning...14 3.0 MEÏH0D0IGGY 3.1 Introduction... 17 3.2 Subjects... ■...18 3.3 M a te ria ls...19

3.4 Data c o lle c tio n ... 20

3.5 V a ria b le s ... . .. .2 2 3.3.1 Dependent V a ria b le ...22 3.3.2 Independent Vai*iable... 22 3.3.3 Control V a ria b le ... 23 3.6 A n alytical Procedui*es... 23 4.0 ANALYSIS OF ÏHS: DATA 4.1 Introduction... 24 4.2 R esu lts... 25

4.3 Discussion o£ the R esu lts... 29

5.0 CONCLUSIONS 5.1 Summary o f the Study...31

5.2 Conclusions... 32

, 5.3 Assessment o f the Study... 33

5.4 Pedagogical Im plications...34

5.5 Im plications fo r Future Research... 35

REFERENCES...37 APPENDICES Appendix A...41 Appendix B... 47 Appendix C... 48 Appendix D . . . ... 49 VI

LIST OF TABLES

TABLE PAGE

3.1 The Matched Scores o f Subjects in the Three

Groups... 19 4.1 Results o f the Test o f Reading Comprehension

fo r Experimental Group 1; Teacher-Generated

Questions... 25

4.2 Results o f the Test o f Reading Comprehension fo r Expei'imental Group 2: Student-Generated Questions...26 4.3 Results o f the Test o f . Reading

Comprehension fo r Control Group... 26 4.4 Means and Standard Deviations o f the Three

Gi'oups... 27 4.6 Results o f One-Way Anova...27 4.6 The Results o f the Analysis o f

Teacher-Generated Questions Answered

Correctly on the T est... ... ... 28 4.7 The Results o f the Analysis o f

Student-Generated Questions Answered

Correctly on the T est... 29

ACKNOWLEDGEMENTS

I v/ould lik e to express my deepest gratitude to

Dr. Eileen Walter fo r her invaluable help and

patience.

I owe special thanks to Dr. James C. Stalker and Dr. Lionel Kaufman fo r th e ir support and guidance.

F in a lly , I wish to thank the BUSEL adm inistrators, and teachez’^s fo r th e ir kindness and cooperation.

ABSTRACT

The goal o f th is study was to determine the

b e n e fits o f prereading a c t iv it ie s in reading

comprehension performance. In th is study the focus

was on, pi’oquestioning, and a comparison o f teacher- genei*ated questions and student-genei*ated questions was raade. I t has been argued in the lite r a t u re that

pi'equestioning has a p ositiv e e ffe c t on reading

comprehension and questioning by students increases tfie e ffe c t . This study aimed to fin d out whether teacher-gexierated or student-generated questioning was moi*e b e n e fic ia l fo r EFL learn ers.

F irs t o f a l l , the theory, research, and methods

on pi*ereading a c t iv it ie s were examined in the

lite ra tu re , and prequestioning was chosen as one o f

tile most e ffe c tiv e ways to improve reading

compi‘ehension. Tlien, the e ffe c t s o f two

prequestioning treatments were studied; teacher­

generated and student-generated questions were

administered to two experimental groups while no

treatment was given to a contx*ol group. A to ta l o f

42 students at Bilkent U niversity (Ankara, Turkey)

participated in the study. Tlie i*esults o f the three gi‘Oups were compared using a One-Way Anova.

The re su lts were inconclusive. Although there was

no sign ificaxit differen ce between the groups, there

direction expected, in dicating that student—generated q^uestloning seemed to improve comprehension more than

teacher-generated cLuestioning and more than no

questioning. Further research with a la rg e r number o f iects is i*ecommended.

CHAPTER I INTRODUCTION

1,1 Background o f the study.

Goodman has described reading as "a

psycholinguistic guessing game, in which the reader reconstructs as best as he can, a message which has been encoded by a writex* as a graphic display" (Gioodman, 1971, p. 135). Goodman views th is act o f coiistruction o f meaning as a c y c lic a l process o f sajiiplirig from the input te x t, predicting, te stin g and confirming those predictions, and sampling fu rth er. In th is model, the reader need not use a l l . o f the

clues. Reading is a receptive language process. I t

s ta rts with a lin g u is t ic surface representation encoded by a w rite r and ends with the meaning which

the reader constructs. There i s an e sse n tia l

interaction between language and thought in reading. The w rite r encodes thought as languaige and the reader decodes language to thought.

P ro fic ie n t readers are both e ff ic ie n t and e ffe c tiv e . They ax*e e ffe c tiv e in constructing a meaning that they can assim ilate. The construction o f meaning bears the same le v e l o f agreement with the o rig in a l meaning o f the author. E ffic ie n t readers oiinimize dependence on v is u a l d e t a il. The better the reader maikes correct predictions, the le s s v is u a l perceptual infoi*mation the reader I'equires.

AtitMiX'diïiii to Goodîiiaxi,

... the reader does not use a l l the

information a v a ila b le to him. Reading i s a process in which the reader picks and chooses from the a v a ila b le information only en ou ^ to aelvect and predict a language structure which

is decodable. I t is not in any sense a

pi'ecise perceptual pi*ooess. (1973, p. 164)

C a rro ll (1988) noted' that the immediate goal o f EFL reading teachers i s to minimize reading d i f f i c u l t i e s

ami to maximize compi*ehension by providing related

information- Goodman (1979) put the issue into focus

when lie said tliat

oven highly e ffe c tiv e z'eaders are severely limited in compi»ehension o f texts by what they ali^eady know befcxre they i-ead. The author may

influence the comprehensibilty o f a , text

p a rtic u la rly fo r s p e c ific targeted audiences. But no axjthoi^ Ccua completely compensate in v/riting fo r the range o f d ifferen ces among a l l potential z*eaders o f a given text. (p . 658)

Rivers and Tempexd.y (1978) emphazised the importance

of providing background information, explaining high-r:t‘e<iuency words, and using illu s t r a t io n s with reading passages to provide addition al meaning to the texts. Coady (1979) suggested that background Imowledge might fuî'ther be able to compensate fo r certain syntactic d e fic ie n c ie s :

Ihe subject o f x^eading m aterials should be

high in terest and re la te w ell to the

background o f the reader, since strong

semantic input can help compensate when

syntactic control is weak, the in terest and background knowledge w i l l unable the student

to comprehend at a reasonable rate and keep

him involved in the m aterial in spite o f i t s syntactic d if f i c u lt y . (p . 12)

Carre11 (1988) suggested that since no author can compensate fo r the in dividual vai'iation among readers,

the teacher in the EFL classroom, can approach th is

I>i*oblem by manipulating eith er one o f the two

v a ria b le s : the text and/or the reader.

Providing background information and previewing

content fo r the readei“ seem to be the most important striitegies foz· the langucige teacher (C a r r e ll, 1988).

The teacher' can manipulate the students' background

information in d iffe re n t ways. C a rre ll noted that pi'eviewing i s an important a c t iv ity fo r activatin g background knowledge in the reading classroom, but i t is not n ecessarily a process simply providing a preliminary outline o f a given text. Sometimes i t can involve teaching a key concept which i s c u lt u ra lly loaded, and asking open-ended questions which help the

teacher to fin d out the c u ltu ra l problems that

students themselves brin g to the text. The prereading

exercises are usuaJ.ly in the form o f prediption

questions, or information seeking questions. These

exei'clses get the student to predict what the text w i l l be about. They accomplish both go als: bu ild in g

new backgroimd knowledge as w e ll as · activatin g

(ixis ting knowledge.

Thus, in achieving the goal o f reading

comprehension in the EFL classroom, the balance

texts the students read and the background knowledge the students possess should be considered.

1.2 Goal and Ratiomile o f the Study

■^rhe .· purpose o f th is study i s to determine the

e ffe c t o f prereading a c t iv it ie s on reading

comprehension performance o f EFL students.

S p e c ific a lly , th is study aims to determine the

relation sh ip between the prequestions asked by

teachers and pi*equestions asked by students and the reading compi‘ehension performance o f EBIj students as iTieasured on a m ultiple choice exam.

In order fo r students to complete th e ir u n iversity studies, I'eading is a necessary s k i l l . This s k i l l

should be einphazised when students study at the

uxiiversity. W iriyachitra (1986, p. 148) reported that "our students must demonstrate an acceptable le v e l o f

understanding o f m aterials designed fo r native

speakers, e . g . , p o lit ic a l science, bio lo gy , etc. Thus, the essen tial s k i l l i s reading although the other s k i l l s , liste n in g and speaking, should not be

ignored."

As researchers have noted (Anderson & Pearson, 1984; Kintsch & Van D ijk , 1978; Wilson & Anderson,

1986) background knowledge has a great impact on

¡"cading comprehension. A great amount o f research has

arfjued that p rio r knowledge o f te x t-re la te d

inl'ormat.ioii sti*oiigly a ffe c t s reading comprehension. Mayner (1984) reported that studies o f prereading a c t iv it ie s have demonstrated the f a c i lit a t i v e e ffe c t s o f activatin g readers' p rio r knowledge relevant to the

compi'ehension o f the new text. Hansen (1981)

suggested that the prereading a c t iv it ie s both prepared foi' the new concepts and make the reading task more enjoyable. These a c t iv it ie s also connected the new c<intent 1120re meaningfully to p rio r knowledge.

Schema theory (Rumelhart & O.rtony, 1977) has

explained how prereading a c t iv it ie s improved reading comprehension. Accox'ding to th is theory, meaning i s reconstructed thx‘ough in teraction between the re a d e r's schematic ( i . e . , knowledge structures in телюгу) and

tiic text. Rumelhax't (1981) explained how the

activation o f appx’opriate scheniata helps i^eaders to

uiuiex'stcind text. Comprehension might be impaired

i f tlie i^eader fa ile d to activate the appropriate SidKiffia.

Л lo t o f 1’eaeai‘ch has been done on prereading

a c t iv it ie s . One tliat was done by T aglieber et a l . (1988) showed that there was a p ositiv e re latio n sh ip

between prereading a c t iv it ie s and reading

comprehension. The i*esults o f th is study suggest that prereading a c t iv it ie s might be a usefu l tool fo r teachiix's o f ЕБЪ to prepare students to comprehend what they re<ad.

Px^equesbioning i s one oP the px*ereading techniques bliat seems p ra c tic a l fo r EFL learn ers. Hansen (1981) and Singer & Donlan (1982) reported that students set

purposes through prequestioning and ask questions

v/hose answers requix'e x^eading fo r comprehension.

1.3 Dei in i tions

Schema. or p rio r knowledge, i s the knowledge a readex' bx-ings to a text that is re la te d to the con bent domain o f the text (Cari*ell, 1987).

Prereading a c t iv it ie s are those cairried out before the reading task to activate p rio r knowledge and set a

pux’pose fo r reading. They may involve pictures,

vocabulax*y pi*e teaching, prequestioning, and other

i-iicbivibies.

P roouesbioning is one o f the prereading

a c t iv it ie s . Eibhex* the teacher asks questions fo r ijbudenbs to answer while reading or the students

gonex-ate bheii* own questions to answer. I t may

consist o f givin g students the t i t l e , a picture, or ox'al summai'y o f the reading passage, end asking the students to formulate questions.

1.4 Hypotheses

Expex^imental Hypotheses: There i s a positive

xuilationship between prequestioning by teachers end

by st;udents and the reading comprehension o f EFL

student generated q.uestions w i l l have a greater e ffe c t than teachez* generated questions.

Null Hypothesis: Tliere i s no relation sh ip between

prequestioning by teachers and by students and the

x'tiading compreliension o f EFL students as measured on a multiple choice exam.

1.5 V ariables

Dependent v a ria b le : Reading comprehension

performance as measured on a m ultiple choice exam. Independent v a ria b le ; Pi'equestioning.

Moderator v a ria b le : Teachers' cuid students'

questions.

Control v a ria b le : Upper-intermediate le v e l

students as measured by the progress te st.

1.6 Ovei'view o f the Me thodology

This study was based on a previous study but

d iffe z‘ently focused. Tiie previous study was done by

Taglieber et a l. (1988). Their study d ealt with

thz*ee prereading a c t iv it ie s and examined the

z'elationship among those techniques and reading

comprehension- In th is study, the focus was on

prequestioning only. In addition, the re s u lt s were calculated d iffe r e n t ly from the Taglieber et a l . ivtudy. TJiey used Manova to te st the re s u lts . In th is study one-way Anova was used.

1. .6.1 Sub J ecbs

There were two experimental ti*eatments conducted

on two groups o f subjects o f the same English

liciency le v e l. The subjects were at the upper- intermediate le v e l o f English as a foreign language as

measured by a progi*ess te st. There were fourteen

subjects in both experimental and control groups. The data were collected at BUSEL (Bikent U niversity School 0.1 English Language), Ankara, Turkey.

1.6.2 Data C ollection

In the f i r s t treatment, prequestions were asked t)y tlie teacher about the general tlieme o f the passage iind the topic. In the second treatment, prequestions

were asked by subjects. The treatment consisted o f

giving subjects a sentence that was c ru c ia l fo r

understanding the theme o f the reading passage. . The students were asked to formulate some questions that the i-'assage might answer*. Tlie topic o f the reading passage was also used as a means fo r generating questions. In the control group a warm-up a c t iv ity

was given. The students were asked i f they liked

reading, and i f not what caused that, and whether

t-hey x'ead a lo t or not. They were also asked to discuss t}ie kinds o f short s to rie s they lik e d . The same amount o f time was used in each treatment and

/'i'oup. Tiion the subjects in each group were given ri.i'l;e(in minutes to i*ead a 1426 woi*d passage s ile n t ly . Ai'terv/ards they were given fift e e n minutes to complete

a test o f comprehension- This te st consisted o f

fifte e n multiple-cho.ice items.

1.7 Data Analysis

'Ihe re su lts o f each group were calculated and compared using a one-way Anova. X>'irst, the re su lts o f teacher-prequestions and then the student-prequestions w(;re compared, and then the re su lts o f each group were (ioinpared with the control grou p's.

I . 8 Exi>ecta tions

The subjects in the student-prequestioning group

wt;re expected t.o perfoz^m better* than the other two

groups. I t was thought that the students, by

activatin g th eir p rio r knowledge, would gain a b etter understanding o f the reading passage, because readers are more successful i f they are a c tiv e ly involved in

tlie reading process. The students activate th e ir

schemata, or p rio r knowledge, eith er by being

questit>ned or by th eir own questions. I t was also expected t,hat students would be most lik e ly to take more e ffo r t to understand what they read when they geruirnte th e ir owti questions. The lowest performance

1.9 L Lniibat ions

Tills study was lim ited to the reading s k i l l in EFL

situ ation s, s p e c ific a lly prequestioning as a

prereading a c t iv ity . Since the main purpose o f BUSEL is to prepare students to study in an English-medium un iversity, i t is d ire c tly applicable to BÜSEL's

pui'pose and i t can be applicable to language

pi*eparatory progz*ams and other im iv e rsitie s with the same purpose.

1.10 Oi’ganisation o f Thesis

The second chapter reviews lite ra t u re on

prereading a c t iv it ie s . The th ird chapter describes the ti'eatment: subjects, m aterials, data c o lle c tio n ,

and an alysis procedures. Chapter four presents the

re su lts o f the te sts which assessed reading

comprehension prof iciexicy. The re su lts o f the

s t a t is t i c a l te sts are discussed in th is section. The la s t chaptei·^ includes a summary o f the study, the conclusions, and th eir application to BUSEL and other EFL situ atio n s. Further research i s suggested.

CHAPTER I I

REVIEW OF LITERATURE

2.1 Inbroducbion

Pr.ioi‘ knowledge that f a c ili t a t e s comprehension has been studied under the t i t l e o f schema theory. As Adams and C o llin s (1979) and Rumelhart (1977, 1981)

noted, schema theory describes the ro le o f pre­

e x istin g knowledge in the process o f reading. Scemata

cU‘e p re -e x istin g knowledge structures, stored

hierai'chiccilly in the brain , the more general,

subsuming the more s p e c ific . Each re a d e r's hierarchy o f schemata organizes her/his knowledge o f language

and the woz*ld. While reading one foi*ms expectations

biised on p rio r knowledge o f texts and the world.

In th is chapter, the ro le o f schema and p rio r knowledge on I’eading comprehension w i l l be discussed.

Ihe research about prereading a c t iv it ie s and

prequestioning w i l l be reviewed in th is chapter.

2.2 P rio r Knowledge

Texts become e a sie r to read i f they correspond to student's pz*ioi’ knowledge o f language and the world.

As Hudson (1982) noted background knowledge has a

s ig n ific a n t impact on the in terpretation o f the texts.

He showed how schemata could override language

proficiency as a fa c to r in comprehension. Smith

process o f reading and concluded that reading i s based on more than the use o f decoding s k i l l s , such as phonics. I f the reader i s not able to connect the new infox'mation in a text to what he or she already knows, then ineaningfu.1 reading w i l l not take p lac e,

I f reading comprehension depends on what the

readex’ already knows, then teachers should c a re fu lly

coi'isidor students' prior* knowledge when choosing

texts- As Nubtall (1982) claimed, reading

comprehension dei>ends on the successful match between re a d e r's p rio r knowledge and the background knowledge assumed by a text. As Barnett (1989) and C a rre ll cind i'’,istex*hold (1983) noted, the reader constructs the

text information based both on the knowledge drawn

from the text and from the p rio r knowledge a v a ila b le to the z'eader.

Garnham (1985), Kintsch (1988), and Rayner and

PoLlatsek (1989) reported that schema theory might not

I>e a w ell-d e fin e d framework fo r the mental

rcipx*esentation o f knowledge but i t has been extremely

useful fox* describing how p rio r knowledge is

integrated in memory and used in higher le v e l

coiaprehension processes. Anderson and Pearson (1984)

noted thcit schema theory can be very useful in

suggesting how to improve reading in struction .

Schemex, or pi'ior knowledge, plays an important

role in prereading a c t iv it ie s . C a rre ll (1987)

iuui F'ioyd and Car.reil (1987) noted that schema theory provides a strong ratio n ale £or prereading a c t iv it ie s . OUhez* I'csecU’cii on schema theoi*y has argued that a high degree o f px*ior knowledge could overcome lin g u is t ic d efic ien c ies ( e . g . , Hudson, 1982). Barnett (1989), Caj-rell (1988), and Dubin and Bycania (1991) implied tluil; students should activate p rio r knowledge o f a iorvi.c before they begin to read. I f the students do not have s u ffic ie n t p rio r knowledge, at le a s t minimal

l>€-ickgx-ound knowledge should be provided. This

knowledge v ;lll help them to in tei‘pz*et the text at some level o f understanding.

2.3 Prereading A c tiv it ie s

I t has been shown tliat prereading a c t iv it ie s pi'oduce s ig n ific a n t ly higher reading performance. For

oxcuiiple, Tagliebei* et a l. (1988) did research on

thr^ee types o f prei'eading a c t iv it ie s : p ic to ra l

Cinu/Cxt, vocabulai*y preteaching, and prequestioning.

A positive I'elationship between these a c t iv it ie s and rciading compi'eiiension performance was obtained.

D iffere n t prei’eading a c t iv it ie s may be more or le ss e ffe c tiv e fo r readers with d iffe re n t proficiency

Jevels. In one study, Hudson (1982) found that

prereading a c t iv it ie s designed to activate students'

scJierna rather than vocabulary preteaching had a

greater e ffe c t on reading comprehension. But at the

fuivanced le v e l iieither o f these types o f prereading act-ivities were any b etter than the other.

Some existin g second language m aterials include

preraadi.ng exercises, svich as information-seeking or prediction questions fo r the reader to keep in mind while reading (e . g . , G re lle t, 1981). A llen and

Widdov/son (1974) noted that question-posing,

prediction, and other prereading exercises supposedly

function to motivate students to read. They also

function to get students to predict, within a general content area, what the text w i l l be about.

2.4 Prequest i oning

Prequestions have been shown to f a c i l i t a t e reading

(.:ompreliension. Leki (1986) noted in such exercises

the teacher slsould deteimiine two or fouz' important issues that require students to seai'ch through th e ir own liv e s fo r sim ilai' expei‘iences. The questions

v/ould not be elaborated. The purpose i s simply to

ivriggor within each student a set o f associations.

Prequestioning can be directed in two ways:

reader-gene j;‘a ted questions and teacher-genei*^ated

questions. Henz'^y (1984) proposed a strategy fo r

i.inpz'oving reading comprehension. Again the re a d e r's prior·^ knowledge takes center stage. The readers have di.fi'ei^eni, ways to accomplish . reading comprehension.

Tteader-generabed questions can help the readers to test th e ir comprehension. The tra d itio n a l question

approach lim its students' comprehension and the

iitiidenbs comprehend in a d iffe re n t manner and more

naturiilly when the questions are not imposed by

O l l i e r s . Student-genera ted questioning, on the other

hand, would avoid th is problem.

There are several techniques that are used in

studenb-generated quesbioning. Henry (1984) proposed

tliree techniques that can be used to stimulate

student-questions: sentence stimulus, thematic

stimulus, and picture stimulus.

Another pi'equestioning type i s teacher-generated

questions. For example, background information can be

provided foi> the students through a p retest. As

Pressy (192G) and Hartly and Davies (1976) found,

pretests increased students' s e n s itiv ity to content and consequently students comprehended more when they read. Such pretests can be in the form o f m u ltiple- choi-ce and true fa ls e questions.

Tlie use o f prequestions makes reading an active

process. As Andre and Anderson (1978-1979) remarked,

b y prequestioning students set purposes fo r reading.

In iiddition to that, the students read to answer th e ir questions or the teachers' questions, and they think ahead o f time what the p ossible answers are. Devine

a c tiv ity in reading and encourages both student- teacher and student-student interaction.

CHAPTER i l l MKTHODOT.OGY

3 . 1 Tr) brod'oc L io n

T>)a main conoer'n o f t h i s tvh aptor i s rriC)'i>ocIo l o g y . I t g i v e s a debni.l.od e x p la n a t io n , o f th e procodi.n.’e s o f t fie s t u d y , .in c lu d in g t lie s u b j e c t s and dai;a c o l l e c t i o n . The fo c u s o f t h i s s tu d y i s on p.re.r eatiiiiE.· a c t i v i t . ie s . As m e n tio n e d foe..foi:'e in Cliapte.?.“ One, a c t i v a t i n g s tu d e n t-s ' p.ni(.)r knov/ledge o r schema h as an im p o rtc in t im p a c t on r e a d in g com preh en sion , o f s t u d e n t s , and p > re -< iu e s tio n in g i s a p r e - r e a d i n g a c t i v i t y t h a t can a c t i v a t e sch em a.

A lo t o f research has been done in the f i e l d o f

prequestioning, Cohen (1983) reported that elementary

students' learning increased s ig n ific a n t ly when

students generated the>ir own g[uestions, when compared

to a control group using only teacher-generated

questions. Performance was measured in both a

c r it e r io n -r e fe i’^enced te st and a standardized te st o f achievement.

Various reasons have been given fo r why students

achieve comprehension when they generate th e ir own

questions. Singei* (1978) reported that student­

generated questions allow students to guide th e ir

thinking. In addition, Hansell (1976) suggested that

participabci wi.t/aout the fe e lin g o f fz’usbration oi· fa ilu r e . Vacca (1901) stated that i f students had ci,irioi3ity, then they would have greater motivation to

read. Ti)is cuiviousity can be accomplished through

sbudet\L-genei‘abed questions. One c o n flic tin g study was reported by Bernstein (1973). Black sixth graders

v;ere divided into three groups: generate own

quest, i Oils, answer peer questions, and answer teacher quosld ofis. A fter students read tlie passage, they were given a m ultiple-choice test. Tlie re su lts indicated Lriat a l l groups performed the sfune. However, the giviup i;‘equir“ed lio genez*^ate questions had not been aot.lviily taugiit quest ion-genera ting s tra te g ie s.

3 .2 SubJ e c 1-s

F.FL subjects were selected by the researcher at

the BUSEL (Bilkent University School o f English

rjc-anguage) wltli the permission o f the directo r o f

BiJGEL. Based on the scores o f a progress test

recently taken in retiding comprehension the means o f

eight classes were compax*ed, and three classes with

nimiliii· nieans wez'e chosen foz' the treatment. The

subjects In each group wez*e matched according to th eir scores. The lowest score was 10, and the highest one was 10 out o f 25.

'i’hei*e wez*e fourteen subjects in each gz*oup. Tills wat·; due to the absence o f some students and mismatch of some scioz'es. The age range wгıs 19-21 in the three

groups. A ll o f tliem wei*e at the L6 le v e l. According to leveJs o f proficiency at BUSEL, th is i s upper- intermediate. The to ta l number of the subjects in the treatment was 42. Tlie z*esults o f the matching are siiown iii Table 3.1.

Table 3.1

The Matched Scoi*es o f Subjects in the Three Groups

Subjects Group I Group I I Group I I I

1 10 10 10 2 11 11 11 3 12 12 12 4 12 12 12 5 12 12 12 6 14 14 14 7 14 14 14 0 15 15 15 9 16 16 16 10 16 16 16 11 16 16 16 12 17 17 17 13 17 17 17 14 18 18 18 3.3 M aterials

The m aterials used in th is study irxcluded a short story o f 1425 words, two worksheets, and a m u ltiple- clioice te st.

The shoi't story was appropi^iate to the le v e l o f

the subjects (see Appendix A ). The short story was

tak<r>n from a co lle c tio n o f s to rie s used fo r upper-

iniiermediate le v e l students (Alptekin and Gürel,

1990). Tlij.s c o llec tio n o f s to rie s i s aimed fo r use

hy lAx/iXir-itil-ermediate students. The theme o f the

selected story appeals to students' in terests and

l i f e . I t is about a teen-age boy captured by the

(‘asiiion o f the day. He is so influenced by

iiiaterialistic thiags to such an extent that he can not prevent himseri: from ste a lin g things.

Thfi worksheets were prepared by the researcher.

The purpose; o f the worksheets was to activate the

biiokgi’fAund knowledge o f the subjects. For the

teaciit;r-gt;aera t.ed que;sbion group, the worksheet

(;ons i st(;d o f cpieistions that were about the general tlie;me of the stoi*y; the questions preceded a short

poem (see Appeindix B ). For the student-generated

question gt'oup, the workstieet consisted o f a sentence tliot was cruc;:jal to the theme o f the story (see Apx>endix C).

The test was also j>repared by the researcher to measure the reading compi*ehension o f the subjects (see

Appendix. D) - I t was in m ultiple-choice format, and i t

consisted o f fifte e n questions. The use o f a

m ultiple-choice exam was preferred because i t could be

taken in a short amount o f time, students were

fa m ilia r with the format, and i t was easy to score.

b.4 Data C ollection

In the f i r s t experimental group, teacher-generated

<luest.j.ons were given to the subjects by the

foseai'ciier, and the subjects answered those questions, ¡•’i r s t of a l l , the researcher asked some o ra l questions

aboub the topic o f a given stoz*y. These questions

included who, why, and where type questions. The

subjects had various answers fo r these questions,

oeiiondly, the students wei*e given a worksheet

c;t)m>isting of teachez'-generated questions. The

students, while answez'ing those questions made

associations from th e ir own liv e s , and in th is way they were much involved in the siibject matter. They

were given fifte e n minutes to complete th ese.

a c t iv it ie s . Tlien the reading passage was d istribu ted , and the subjects wez*e given fifte e n minutes to read i t ;>i 'icntly. F in a lly , a m ultiple-choice test was given

to i.hem to complete in fift e e n minutes.

Ixi the student-genez*ated questioning group, f i r s t o f a l l , the subjects were told to generate some

quesl,ions about the topic o f the given story. The

students discussed i t among themselves, made guesses

eibout wliat questions this story might answer.

Af tez’v^az’d s, the students were given a worksheet with a sentence about the topic and generated again th e ir own questions T'elated l;o the story. Tliese a c t iv it ie s also took the same amount o f time as the previous group. The scuae z'eading passage was given to th is group. Tliey wer^e also allowed the same amount o f time to read and answez‘ the same m ultiple-choice te st.

The control group was given a fift e e n minute warm­ up a c t iv ity instead o f preqiiestioning. They were questioned <ibout th e ir hobbies. The focus was on reading. The students expressed th eir fe e lin g s about reading d iffe re n t kinds o f short sto rie s (scien ce- fic tio n , adventure e t c . ) . In the second and third part o f the tz'e<itinent the same procedure was applied. They also read the same passage and answez»ed the same niultii^le-choice test with same time lim itation as the otiler two groups.

In th is research p ro ject, the relation sh ips

between teacher-generated and student-generated

questions and reading compz-ehension were investigated,

in p a rtic u la r, whether teacher-generated and/or

iitudent-generiited questions were b e n e fic ia l in reading oompz-ehension performance.

3.5 V ill' i ab 1 e s

3.3.1 Dependent Viiz'iable

The I’^eading comprehension o f the subjects

mcasuj'ed on a m ultiple-choice exam was the dependent

vaz-iable in th is study. Subjects” reading

com prehension depended upon the type o f treatment and

no t.rea bment.

3.3.2 Independent Vai'iable

The independent v a ria b le was prequestioning.

or the ijbudenta as moderator v a ria b le s. 3.3.3 Con tro1 Variable

The subjects proficiency le v e l that was

previously detei'niined by the progress te st at BUSEL

was coxitrolled. Tiie subjects were at the

upper-in torupper-inediate le v e l .

3.6 A nalytical Procedures

The test re su lts o f each group were calculated and compared using a one-way Anova. F ir s t, the re su lts o f student-prequestions and the teacher-prequestions were compared, and then the re su lts o f each group were, compared with the control grou p's.

CPIAPTER IV

ANALYSIS OF THE DATA

4.1 Introduction

In th is study, i t i s argued that prereading a c t iv it ie s should be used to f a c i li t a t e reading comprehension. That is , i f prereading a c t iv it ie s are selected which activate th e ir pi’^ior knowledge and set a pxirpose foi' reading, students can gain a b etter understanding o f a reading passage.

In ordei* to understand the ro le o f prereading a c t iv it ie s in EFL c la sse s, the reading comprehension

o f students at the preparatory school o f Bilkent

Uiilvcirsity (Ankara, Turkey) wex’e studied. Three

groups o f subjects were given a prereading treatment. In the f i r s t gi'oup, teacher-generated questions, in the second one student-generated questions, and in the contixil groujj wai'm-up a c t iv it ie s were accomplished. Then, the studexits were given a short story to read, followed by a lb m ultiple-choice te st to measure th eir reading comprehension.

It was hypothesized that there would be a p ositive relation sh ip between prequestioning by teachei's and

by students cind the reading comprehension o f EFL

students as measured on a m ultiple-choice excim and

student-generated questions would have a greater

4.2 Results

In the f i r s t experimental group, teacher-questions were asked as a px'ex‘eading a c t iv ity . The re su lts o f the te st o f reeiding comprehension fo r th is group i s given in Table 4.1.

in the s<icond expei’imexital group, student-

questions were genex'ated and discussed. The re su lts of the test o f x'eading compx'ehension fo r th is group are s'hov;n in Table 4.2.

lii the control group, warm-up a c t iv it ie s were })i*ovided befox'e the students x*ead. TTae re su lts o f the test o f i-eading compi'ehension fo r control group are shown in Table 4.3.

Table 4.1

Results of the Test o f Reading Comprehension

fo r Experimental Gi’oup 1: Teacher-Generated Questions

e>c ts liaw Scores Percentages

i 10 G6.6 2 5 33.3 3 5 33.3 4 2 13.3 5 9 60 6 8 40 7 12 80 8 13 86.6 9 12 80 10 11 73.3 11 12 80 12 7 46.6 13 10 66.6 14 12 80 25

Table 4.2

ResuJttj o f the Test o f Reading Comprehension fo r E>:perimental Group 2: Student-Generated Questions

ilubj ects Raw Scores Percentages

1 9 60 2 8 53.3 3 9 60 4 5 33.3 5 7 46.6 6 6 40 7 11 73.3 8 11 73.3 9 11 73.3 10 11 73.3 11 14 93.3 12 4 26.6 13 13 86.6 14 12 80 Table 4 .3

Resxilts of the Test o f Reading Comprehension fo r Control Group

Subjects Raw Scores Percentages

1 10 66.6 2 10 66.6 3 7 46.6 4 10 66.6 5 7 46.6 8 7 46.6 7 7 46.6 8 5 33.3 9 6 40 10 10 66.6 11 6 40 12 6 40 13 9 60 14 10 66.6

ITie means o f the three groups were compared using a une-v/ay Anovii. The means euid standard deviations o f

Table 4.4.

Means and Standai'd Deviations o f the Three Groups

Gx'oup Mean SD

J . Teaciier-que3tio.Tis 69.9 22.8

- Student-queBtion 62.2 20.1

3 .Control 52.5 12.6

■'i’he i*esulbs o f Lhe s t a t is t ic a l an alysis are shown in

Table 4.5. The re su lts o f the two experimental

/»roups, teacher-generated and student-generated

questions, and the control group were compared. I t

v/as found that there was not a sig n ific a n t

differeijce bel,ween the groups so the n u ll hypothesis

is accepted. Howevei*, quest ion-generation may have

contributed to greater variance in the scores since

l.he variance o f the teacher-generated questions is

22.8, student-generated ones i s 20.1 while that o f control gx'oup is 12.6.

Table 4.5

Results o f One-way Anova

Source o f Variance Ss d . f M.S

Hetv/een Groups Wi. thin Groups Total 718.9 14113.2 14832.1 2 39 41 359.4 361.8 .9933

In the te st o f reading comprehension some o f the questions re fle c te d eith er the teacher-generated or student-generated questions asked during the pre-

roxiding stage. Five o f the teacher questions and

live o f the student questions were in the te st.

Subjecfca' answers were analyzed to determine how many o f tiui teacher-generated questions in the Experimental

Group 1, and student-generated questions in the

Expel'imental Group 2 were correct. Ihen, the rate o f success on teacher questions wei'e compared to the rate o f success on sbiident questions. Tlie re su lts o f th is analysis also do not show a s ig n ific a n t d ifferen ce

betv/een the ti'eatment groups. However, Group 2

(student-questions) had a higher mean than Group 1

( tecicher-questions). The mean o f student—questions

wixs 67.1, and the mean o f teacher-questions was 60. The I'esults o f th is an alysis are shown in Table 4.6 and 4 . 7 .

Table 4.6

The Results o f the Analysis o f

Teacher-Genez'ated Questions (n -5 ) Answered Correctly in the Test

SubJects Number o f items Percentages

1 4 80 2 3 60 3 3 60 4 0 0 5 2 40 6 2 40 7 4 80 8 5 100 .9 4 80 10 3 60 11 2 40 12 2 40 13 4 80 14 4 80 M-60 SD-26.02 28

s

Table 4.7

Tlie Results o f the Analysis o f tudent-Generated Questions (n=5)

CoiU'ectly in the Test

Answered

Subj ec ts Number o f items Percentages

1 kJ 60 2,. 4 80 3 3 60 4 2 40 5 3 60 6 2 40 7 4 80 0 4 80 9 4 80 10 3 60 11 5 100 12 3 60 1 3 60 14 4 80 M-:G7.1 Si)·--16.84 •1.3 Diaeusaion o f Results

The re su lts o f the s t a t is t ic a l te sts show . that there i s no s ig n ific a n t d ifferen ce between the groups

ixv terms o f teacher-generated and student-generated

questions. Thei'efore the n u ll hypothesis o f no

differen ce between teaclier-generated and student­

generated questions i s accepted. Although the

s t a t is t ic a l resu lts indicate no sig n ific a n t relationship between the gi'oups, in fa c t, there seems to be sonie indiccition o f a d ifferen ce between the

groups. Group I I (student-questions) achieved higher

and Groiip I I achieved higher re su lts than the Control

Group as was expected before the application o f

treatments.

CHAPTER V CONCLUSIONS

5.1 Summary o i the Study

Prereadiiig a c t iv it ie s which activate students' prior·· knowledge have been Pound to improve students'

reading comprehension. Tliex‘efore, language teaching

auldioritiea empiiasize the importance o£ these

a c t iv it ie s in the EFL classroom, and prequestioning

is one o f tliese prereading techniques. Students

benefit from prequestioning by activatin g th e ir

background knowledge, which helps them to perform

bettt>r on measures o f reading comprehension.

'riiis study aimed to fin d out i f there was a

ixjsitive relation sh ip between prequestioning by

Iveachers and students and the reading compi^ehension o f

the students. Tlie prequestioning treatment consisted

o f two groups: teacher-generated questions and

student-generated questions. A control group was not

givc-in prequestions. Each o f the three groups was

given eltliez' one variety o f the treatment or no

tz^eatiaent, the passage to read, and then a m ultiple-

choice te st to measui’e reading comprehension. Tlie

gi‘oups, which were at the same language proficiency

le v e l, read the same passage and took the same test under the same conditions.

5.2 Coneluaions

The an alysis o f bhe data collected in th is study that both teacher-generated and student-

HOiieratcd questions re su lt in the same reading

(iOMiprehension pei’forniance. Thei‘e was not a

s ig n ific a n t differen ce between the two treatments.

Willie i t cannot be concluded that prequestioning

sign.i ^^^¡¿int].y inci’eases students' reading

coinprehension. The re su lts o f th is study suggest

tliat pi’cquestloning can f a c ilit a t e treading

coniprchensioii. Both o f the experimental groups did

thiin the control group on the te st o f reading

coniprohonsion. In addition, there i s an indication

thai. student-generated questions are superior to

!i(ifualu;r-genci'ated questions. Although the re su lts may be; (iuo to chance, they are in the direction found in pr(;v'ic)us research, lending support to the theory that fu.· i/. vation o f i:>rior knowledge f a c i li t a t e s reading i i( nnprehens.i. oil.

i t cannot be denieei that any pre-reading a c t iv ity

that activates p rio r knowledge, motivates students,

cind focuses th e ir attention and/or sets a purpose fo r reading f a c ilit a t e s reiading comprehension. This

Iniporl.anco o f p rio r knowledge on reading comprehension has been argued by many f i r s t language i^searchers (Andei^son & Pearson, 1984; Bransford, Stein, & rjhe ilion, 1984; Kintsch & van D ijk, 1978; Wilson &

Anciersor·., .1986). However, i t remains to be determined v/liethe.r i t helps foreign language learnei's as w ell as i'la t; ive speti kei*s.

5.3 Assessment o f the Study

The x*esults o f the study may have beexx affected by

several fac to rs. The low number o f subjects, the

lack o f cooperation o f the subjects, the tim eliness o f tlui stxxdy c'lxuJ the Icxck o f appropx'iate topic may have produced unx'o..l.:i.able cxnd in valid re su

lts-Fdrat, tlie attendance o f students becomti lower during the c o llec tio n o f data o f th is study. This made i t d i f f i c u l t to match the students into 3 groups.· This, aliso affected the r e l.ia b ilit y o f the study because o.C the low number o f subjects. Second, the subjijcts were nob .fully cooperative. This stems from the general c h a rac teristic s o f pi*ep school students at

u n iv e rsitie s; they sometimes lack motivation,

cispccially I f they have spent more them one year

at prep school. Third, the experiment was carried

out dux'ing spring term emd the attendance o f prep

school students becomes lower at th is time. They

l)ccome relaxed and lose th e ir in terest in the lessons

more than otliei· times o f the year. Fourthly, the

topic and z*eading passage which were chosen were not appropriate fo r generating many questions. The t i t l e

other a c t iv it ie s were done to aid the prequestioning iictivity .

Similai" re su lts were achieved by other researchers due to in su ffic ie n t methodological procedures. Wong (1985) stated that in the in terpretation o f re su lts

one should be very cai’e f u l , since the treatment

pz'ocedui'es could be inappropriate and therefore the resu.lts could be misleading.

5-4 Pedagogical Implications

'llie importance o f prequestioning cannot be

ignored. They ai*e essen tial fo r reading instruction. The students make guesses and inferences about the

text through prequestioning. Heightened p rio r

knowledge plays an important ro le fo r adequate

comprehension to occur.

The students should be encouraged to generate

thoir own questions foi* the ben efit o f a b etter

uiuiersLanding. Allowing students to take an active

role in th e ir learning should increase th e ir

motivation bo read. In addition, student questioning

encourages the reader to focus attention on the

reading m aterial and set purposes fo r reading. Of the Lhree prereading a c t iv it ie s examined in the study, Leaclier-generated questions was le s s e ffe c tiv e than

student-generated questions although superior to

The i‘ole o f other prereadiiig a c t iv it ie s fo r

better compi*ehcnsion cannot be neglected.

Px*equestioning can be mox'e e ffe c t iv e i f i t i s integi’,ated with other prereading a c t iv i t ie s , such as

p ic to ra l context. The techniques should be chosen

iiccoi'ding to the content o f the passages. Abstract

passages would need prequestioning whereas highly concz'ete passages should be introduced with by the aid o f p ic to ra l context (T a g lie b e r et a l . , 1988).

The goal o f teachers should be in the direction o f pi*oviding a wide range o f e ffe c tiv e s tra te g ie s to

f a c i l i t a t e reading comprehension. Clarke and

S iib e rste in (1977) reported that students need to be taught I'eading s tra te g ie s , such as guessing from the context and finding clues to read e ffe c t iv e ly and c f f ic e n t ly .

5.5 Im plications fo r Future Research

Several suggestions can be made fo r future

x'esearch in the use o f prereading a c t iv i t ie s in EFL cliisses in Tui'key. F ir s t , future studies should be

carried out with a higher number o f subjects, which

w i l l give £i more r e lia b le i‘e s u lt. Secondly, the appropriateness o f the t i t l e and the reading passage must be) taken into consideration. The t i t l e o f a given passage is the most e ffe c tiv e quide fo r the